🐦 Twitter Post Details

Viewing enriched Twitter post

@sudoingX

anon if you have a dgx spark, the best model to run in 2026 is stepfun's step 3.7 flash. i called it the day it dropped, and after living with it on the box i'm saying it louder now. what you're looking at is a 198B mixture of experts vision language model, ~11B active per token, Q4_K_S at ~104gb, running on a single 128gb dgx spark under hermes agent. one machine. it holds the full 262K context at ~25 tokens a second and takes image input. i measured that speed myself at full context. reasoning, vision, frontier size, all of it fitting and running locally on one box. people keep asking what's actually worth running on 128gb of unified memory. this is the answer, and has been since launch. if you bought a spark to run frontier models locally instead of renting them, this is the one to put on it.

Media 1

📊 Media Metadata

{
  "media": [
    {
      "url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2068753345842667744/media_0.jpg",
      "media_url": "https://crmoxkoizveukayfjuyo.supabase.co/storage/v1/object/public/media/posts/2068753345842667744/media_0.jpg",
      "type": "photo",
      "filename": "media_0.jpg"
    }
  ],
  "processed_at": "2026-06-22T06:35:19.745141",
  "pipeline_version": "2.0"
}

🔧 Raw API Response

{
  "type": "tweet",
  "id": "2068753345842667744",
  "url": "https://x.com/sudoingX/status/2068753345842667744",
  "twitterUrl": "https://twitter.com/sudoingX/status/2068753345842667744",
  "text": "anon if you have a dgx spark, the best model to run in 2026 is stepfun's step 3.7 flash. i called it the day it dropped, and after living with it on the box i'm saying it louder now.\n\nwhat you're looking at is a 198B mixture of experts vision language model, ~11B active per token, Q4_K_S at ~104gb, running on a single 128gb dgx spark under hermes agent. \n\none machine. it holds the full 262K context at ~25  tokens a second and takes image input. i measured that speed myself at full context.\n\nreasoning, vision, frontier size, all of it fitting and running locally on one box. people keep asking what's actually worth running on 128gb of unified memory. this is the answer, and has been since launch.\n\nif you bought a spark to run frontier models locally instead of renting them, this is the one to put on it.",
  "source": "Twitter for iPhone",
  "retweetCount": 17,
  "replyCount": 22,
  "likeCount": 214,
  "quoteCount": 4,
  "viewCount": 14517,
  "createdAt": "Sun Jun 21 17:50:07 +0000 2026",
  "lang": "en",
  "bookmarkCount": 201,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2068753345842667744",
  "displayTextRange": [
    0,
    274
  ],
  "inReplyToUserId": null,
  "inReplyToUsername": null,
  "author": {
    "type": "user",
    "userName": "sudoingX",
    "url": "https://x.com/sudoingX",
    "twitterUrl": "https://twitter.com/sudoingX",
    "id": "1555661341914198016",
    "name": "Sudo su",
    "isVerified": false,
    "isBlueVerified": true,
    "verifiedType": null,
    "profilePicture": "https://pbs.twimg.com/profile_images/2059155391200833536/szNfExnk_normal.jpg",
    "coverPicture": "https://pbs.twimg.com/profile_banners/1555661341914198016/1780050322",
    "description": "",
    "location": "Bangkok, Thailand",
    "followers": 31380,
    "following": 1022,
    "status": "",
    "canDm": true,
    "canMediaTag": true,
    "createdAt": "Fri Aug 05 21:05:57 +0000 2022",
    "entities": {
      "description": {
        "urls": []
      },
      "url": {}
    },
    "fastFollowersCount": 0,
    "favouritesCount": 7920,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 797,
    "statusesCount": 9327,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [
      "2068011622510260467"
    ],
    "profile_bio": {
      "description": "GPU/local LLM. more RAM and OSS... everywhere",
      "entities": {
        "description": {}
      }
    },
    "isAutomated": false,
    "automatedBy": null
  },
  "extendedEntities": {
    "media": [
      {
        "allow_download_status": {
          "allow_download": true
        },
        "display_url": "pic.twitter.com/LEAQisxLAn",
        "expanded_url": "https://twitter.com/sudoingX/status/2068753345842667744/photo/1",
        "ext_master_playlist_only": [],
        "ext_media_availability": {
          "status": "Available"
        },
        "ext_playlists": [],
        "features": {
          "large": {
            "faces": []
          },
          "orig": {
            "faces": []
          }
        },
        "id_str": "2068750602637873152",
        "indices": [
          275,
          298
        ],
        "media_key": "3_2068750602637873152",
        "media_results": {
          "id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAARy1rbZfGxAACgACHLWwNRLaYOAAAA==",
          "result": {
            "__typename": "ApiMedia",
            "id": "QXBpTWVkaWE6DAABCgABHLWttl8bEAAKAAIctbA1Etpg4AAA",
            "media_key": "3_2068750602637873152"
          }
        },
        "media_url_https": "https://pbs.twimg.com/media/HLWttl8bEAAceBo.jpg",
        "original_info": {
          "focus_rects": [
            {
              "h": 1434,
              "w": 2560,
              "x": 0,
              "y": 0
            },
            {
              "h": 1573,
              "w": 1573,
              "x": 683,
              "y": 0
            },
            {
              "h": 1573,
              "w": 1380,
              "x": 779,
              "y": 0
            },
            {
              "h": 1573,
              "w": 787,
              "x": 1076,
              "y": 0
            },
            {
              "h": 1573,
              "w": 2560,
              "x": 0,
              "y": 0
            }
          ],
          "height": 1573,
          "width": 2560
        },
        "sizes": {
          "large": {
            "h": 1258,
            "w": 2048
          }
        },
        "type": "photo",
        "url": "https://t.co/LEAQisxLAn"
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {
    "hashtags": [],
    "symbols": [],
    "urls": [],
    "user_mentions": []
  },
  "quoted_tweet": null,
  "retweeted_tweet": null,
  "isLimitedReply": false,
  "communityInfo": null,
  "article": null
}